Generative AI Leader Practice Test 12 — 15 Questions

Question 1

What is the primary purpose of the transformer architecture in large language models (LLMs)?

Accepted Answer

To enable parallel processing of tokens and capture long-range dependencies through self-attention. The transformer architecture's primary purpose is to enable parallel processing of all tokens in a sequence while capturing long-range dependencies through its self-attention mechanism. Unlike recurrent neural networks (RNNs) that process tokens sequentially, transformers compute attention scores between every pair of tokens simultaneously, allowing the model to weigh the relevance of distant tokens without the vanishing gradient problem. This parallelization and global context capture are the foundational innovations that make large language models (LLMs) scalable and effective for tasks like text generation and understanding.

Answer

To generate images from text descriptions

Answer

To convert text into numerical embeddings for downstream tasks

Answer

To store and retrieve information from a vector database

Question 2

Which THREE capabilities are provided by Vertex AI Agent Builder? (Choose three.)

Accepted Answer

Integration with Dialogflow CX for conversational flows.. Vertex AI Agent Builder integrates with Dialogflow CX to enable the design of sophisticated conversational flows, including state management, conditional logic, and multi-turn interactions. This allows developers to build agents that can handle complex dialogues with branching paths, leveraging Dialogflow CX's visual flow builder and fulfillment capabilities within the Vertex AI ecosystem.

Answer

Automated model hyperparameter tuning.

Answer

Support for multimodal (text, image, video) input processing in agents.

Question 3

A government agency is deploying a generative AI system to assist in processing visa applications. The system must be transparent about its limitations and provide clear indications of its confidence in each recommendation. Which approach BEST addresses explainability and accountability?

Accepted Answer

Provide confidence scores alongside each recommendation and include disclaimers about the system's limitations. It directly addresses explainability by providing confidence scores, which quantify the model's certainty, and accountability by including disclaimers about system limitations. This aligns with responsible AI principles that require transparency in automated decision-making, especially in high-stakes domains like visa processing where users must understand the system's reliability and potential biases.

Answer

Only allow the system to output binary decisions (approve/reject) to reduce confusion

Answer

Train the model to always output high confidence to reassure decision-makers

Answer

Use a random subset of applications to be reviewed by humans only, without any AI input

Question 4

Which tool from Google's Responsible AI toolkit is designed to document the intended use, performance, and limitations of a machine learning model?

Accepted Answer

Model Cards. Model Cards are the correct tool because they are specifically designed to document the intended use, performance metrics, and limitations of a machine learning model. This standardized documentation format, introduced by Google, provides transparency by detailing evaluation results across different conditions, intended use cases, and known biases, which is essential for responsible AI deployment.

Answer

PAIR Explorables

Answer

Datasheets for Datasets

Answer

People + AI Guidebook

Question 5

What are THREE benefits of using embedding models in a Retrieval Augmented Generation (RAG) system?

Accepted Answer

They compress text into dense vectors for efficient retrieval.. Option A is correct because embedding models transform text into dense, fixed-length vector representations that can be indexed (e.g., in a vector database) and compared via distance metrics like cosine similarity, making retrieval fast and scalable. Option C is correct because embeddings capture semantic meaning, so a query can match documents that are conceptually related even when they share no exact keywords, which is the core advantage over lexical/BM25 keyword search. Option D is correct because a RAG pipeline supplies relevant context at inference time through retrieval, so the generator can answer domain-specific questions without being fine-tuned on that domain's data. Option B is not a benefit of embedding models, since they do not generate training data; that is a separate capability of generative models. Option E is not correct because embedding models produce continuous vectors and retrieval is similarity-based, so outputs are not guaranteed to be deterministic across queries or runs.

Answer

They allow the model to generate new training data automatically.

Answer

They provide deterministic outputs for the same query.

Question 6

A financial services firm uses a fine-tuned model for contract analysis. They observe that the model's performance degrades after a few months because contract language evolves. The team wants to maintain accuracy without full retraining. What is the MOST cost-effective approach?

Accepted Answer

Perform incremental fine-tuning with a small representative sample of new contracts. Incremental fine-tuning updates the existing fine-tuned model's weights using a small, representative sample of recent contracts, adapting to evolving language at a fraction of the cost of full retraining while preserving prior knowledge. This directly addresses drift without the expense of rebuilding from scratch or the inaccuracy of prompt-only fixes.

Answer

Switch to a larger base model and use zero-shot prompting

Answer

Retrain the model from scratch every quarter with all historical data

Answer

Use Vertex AI Model Monitoring to detect drift and alert, then manually adjust prompts

Question 7

Which Google Cloud product provides access to pre-trained foundation models like Gemini?

Accepted Answer

Vertex AI Generative AI Studio. Vertex AI Generative AI Studio is the correct answer because it is the Google Cloud service specifically designed to provide access to pre-trained foundation models like Gemini, allowing users to test, customize, and deploy them via a managed interface. Unlike other services, Generative AI Studio directly integrates with Gemini's API and offers prompt engineering, tuning, and model evaluation capabilities.

Answer

Dataflow

Answer

Cloud Translation

Answer

Vertex AI Model Registry

Question 8

A financial institution wants to use Gemini to analyze customer support transcripts and generate summaries. They need to ensure that personally identifiable information (PII) is not included in the summaries. Which approach should they take?

Accepted Answer

Preprocess the transcripts with Cloud DLP API to redact PII before sending to Gemini. The Cloud Data Loss Prevention (DLP) API provides a purpose-built, scalable service to detect and redact PII from text before it reaches the Gemini model. This ensures that sensitive data is removed at the source, preventing any possibility of leakage in the generated summary, regardless of model behavior or prompt engineering.

Answer

Use a carefully engineered prompt instructing Gemini not to include PII

Answer

Post‑process the generated summaries with a regex filter to remove PII

Answer

Fine‑tune Gemini to avoid generating PII

Question 9

A healthcare startup wants to generate synthetic patient notes for training medical residents. They need the output to follow a strict template with sections: Chief Complaint, History, Assessment, Plan. Which prompt engineering strategy should they use to ensure consistent structure?

Accepted Answer

Provide a few-shot example of the template filled out and instruct the model to follow that format for new cases. Few-shot prompting provides the model with concrete examples of the desired output structure, which is the most reliable way to enforce a strict multi-section template like Chief Complaint, History, Assessment, and Plan. By showing one or more filled-out examples, the model learns the exact section headers, ordering, and formatting conventions to replicate for new cases. This pattern-based conditioning is far more effective than abstract instructions alone for template adherence.

Answer

Use a zero-shot prompt asking for a patient note in bullet points

Answer

Set temperature to 0 and max tokens to a high value

Answer

Use a system prompt that lists the sections as instructions

Question 10

A company is evaluating the ROI of implementing GenAI for code generation. Which metric BEST captures the productivity improvement of developers?

Accepted Answer

Time saved per development task (e.g., from 2 hours to 30 minutes). Time saved per development task directly quantifies productivity improvement by measuring the reduction in effort required to complete a unit of work. If a task that previously took 2 hours now takes 30 minutes, that is a concrete, measurable 75% reduction in time — the most direct and interpretable ROI metric for GenAI code generation. It captures the actual efficiency gain in terms of developer hours, which translates directly to cost savings and capacity for more work.

Answer

Percentage of code that passes unit tests on the first attempt

Answer

Number of lines of code generated per day

Answer

Number of bugs found in production after code review

Question 11

A team is tuning a large language model for a question-answering task. They notice the model gives high confidence scores to answers that are factually incorrect. Which evaluation metric should they primarily use to detect this overconfidence problem?

Accepted Answer

Expected Calibration Error (ECE). Expected Calibration Error (ECE) directly measures the alignment between a model's predicted confidence and its actual accuracy. In this scenario, high confidence on incorrect answers indicates miscalibration, and ECE quantifies this mismatch by binning predictions by confidence and computing the average absolute difference between accuracy and confidence per bin.

Answer

Perplexity

Answer

BLEU score

Answer

ROUGE-L

Question 12

A company is evaluating whether to use a pre-built API or fine-tune a model for their use case. They have a large dataset of domain-specific jargon and need high accuracy on specialized terms. Which factor MOST strongly suggests fine-tuning?

Accepted Answer

The model needs to understand and generate domain-specific jargon accurately. Fine-tuning is most justified when the model must accurately understand and generate domain-specific jargon that a general pre-trained model handles poorly. A large labeled dataset of specialized terms is exactly the signal that fine-tuning will outperform prompt engineering or a pre-built API. The other options describe scenarios favoring pre-built APIs.

Answer

The team wants to rapidly prototype a solution

Answer

The team has a limited budget for compute resources

Answer

The application requires low latency responses

Question 13

Despite applying safety filters, a generative AI model still produces toxic outputs in some cases. Which additional technique should be applied?

Accepted Answer

Use RLHF with human feedback to reduce toxicity. RLHF (Reinforcement Learning from Human Feedback) directly addresses toxicity by using human evaluators to rank model outputs, then fine-tuning the model to prefer less toxic responses. This technique teaches the model to avoid harmful patterns that safety filters might miss, as filters are static and can be bypassed by adversarial prompts or nuanced toxicity.

Answer

Add more examples of toxic content to training

Answer

Increase the filter threshold

Answer

Decrease the model's temperature

Question 14

A data scientist is using Vertex AI generative AI studio to create a chatbot. The chatbot gives inconsistent answers to similar questions. Which parameter should they adjust to make responses more consistent?

Accepted Answer

Decrease temperature to 0.2. Decreasing the temperature to 0.2 reduces the randomness of the model's token sampling, making the output more deterministic and consistent. Temperature controls the probability distribution over tokens; lower values make the model more likely to choose the highest-probability token, reducing variability in responses to similar questions.

Answer

Increase top-p to 0.9

Answer

Increase presence penalty to 0.5

Answer

Decrease frequency penalty to 0.0

Question 15

A company wants to scale their generative AI application globally with low latency. Which infrastructure configuration is most suitable?

Accepted Answer

Multiple regional endpoints with traffic routing to the nearest region.. Deploying multiple regional endpoints with traffic routing to the nearest region minimizes latency by directing user requests to the geographically closest inference endpoint. This architecture leverages global load balancing (e.g., using Anycast DNS or HTTP(S) load balancers with backend services in multiple regions) to reduce round-trip time (RTT) and meet latency SLAs for real-time generative AI applications.

Answer

Use a CDN to cache responses.

Answer

On-premises deployment for all regions.

Answer

Single endpoint in us-central1 with high max replicas.